Discussion.
One of the most important objectives during root-canal instrumentation is the removal of vital and/or necrotic pulp tissue, infected dentine, and dentine debris in order to eliminate most of the microorganisms from the root-canal system (European Society of Endodontology 1994, American Association of Endodontists 1998). The ability to achieve some of these objectives was examined in this investigation on severely curved root canals, involving K3 rotary nickel-titanium instruments and stainless steel hand K-Flexofiles.
Debris was defined as dentine chips, and residual vital or necrotic pulp tissue attached to the root-canal wall which in most cases is infected (Hulsmann et al. 1997).
Thus, debris might prevent the efficient removal of microorganisms from the root-canal system. Moreover, debris may occupy part of the root-canal space, and thus may also prevent complete obturation of the root canal (Wu et al. 2001). The smear layer is a surface film of a thickness of approximately 1-2 mm (American Association of Endodontists1998). Smear layer, which is mainly inorganic, is produced every time a canal is instrumented (Grandini et al. 2002), no smear layer is found on areas that are not instrumented (West et al. 1994). The smear layer contains dentine particles, residual vital or necrotic pulp tissue, bacterial components, protein agglomerates, blood cells as well as retained irrigants, and it blocks the openings of the dentinal tubules (West et al.1994, Grandini et al. 2002). In this way, a thick and non-homogenous smear layer can prevent the efficient elimination of intracanal microorganisms, and compromise the complete sealing of the root canal (Petschelt et al.1987, West et al.1994).
Although it is recommended to use antibacterial irrigants in combination with chelating agents in order to remove debris as well as the inorganic/organic smear layer (West et al.1994, Hulsmann et al. 1997, Gambarini 1999, Grandini et al. 2002), in the present study sodium hypochlorite alone was used as an irrigant. This solution would appear the best available canal irrigant owing to its antibacterial and organic tissue dissolving properties (Spongberg et al. 1973, Turku n & Cengiz 1997), but it is not possible to remove the smear layer with NaOCl (Yamada et al. 1983, Grandini et al. 2002, Guerisoli et al. 2002). Nevertheless, considering the major objective of the present investigation (to solely compare the cleaning effectiveness of the two instrumentation techniques under identical conditions) a simple irrigation technique was used, avoiding any associations of different irrigation solutions. Since it has been recently shown by Hulsmann et al. (2001) and Grandini et al. (2002) that EDTA containing chelating agents (e.g. RC-Prep, Premier, PA, USA, or Glyde File Prep, Dentsply Maillefer, Ballaigues, Switzerland) may be partially responsible for a good cleanliness of the root-canal walls after instrumentation with rotary nickel-titanium instruments, it has to be taken into consideration that the cleaning efficiency of the two instrumentation techniques evaluated in the present study might be further improved using a combination of NaOCl and EDTA containing chelating agent.
In the present study, the cleaning efficacy of two instrumentation methods was examined on the basis of a separate numerical evaluation scheme for debris and smear layer, by means of an SEM-evaluation of the coronal, the middle and the apical portions of the canals (Mizrahi et al. 1975, Bolanos & Jensen 1980, Haikel&Allemann 1988, Hulsmann et al.1997).With both instrumentation techniques, partially uninstrumented areas with remaining debris were found in all canal sections. This finding has also been described by other authors (Bolanos & Jensen 1980, Schwarze & Geurtsen 1996, Hulsmann et al. 1997). Moreover, the present results indicate that on average the apical third of the canals was less clean than the middle and coronal thirds regardless of the instrument used. This observation is also in agreement with other studies (Wu & Wesselink1 995, Hulsmann et al.1997, Schafer & Zapke 2000, Hulsmann et al. 2001, Gambarini & Laszkiewicz 2002).
In general, the use of stainless steel K-Flexofiles resulted in significantly less remaining debris (Table 2) compared to canal shaping with rotary nickel-titanium K3 instruments, whereas for smear layer no significant differences between these instruments occurred (Table 3, P = 0.329). Whilst these results corroborate a previous report, in that stainless steel hand instruments proved to be superior to ProFile rotary nickel-titanium instruments as far as cleaning efficacy is concerned (Schafer & Zapke 2000), in other studies no significant differences between the cleaning efficacy of stainless steel hand and rotary nickel-titanium instruments (Lightspeed, ProFile .04, Rotofiles) were observed (Kochis et al. 1998, Bechelli et al. 1999, Hulsmann 2000). More recently it has been shown that some other modern rotary nickel-titanium instruments, such as Hero 642 and Quantec SC instruments (Hulsmann et al. 2001), GT rotary files (Gambarini & Laszkiewicz 2002) or FlexMaster instruments (Schafer & Lohmann 2002) showed a good cleaning ability. Generally, the canals prepared with these rotary instruments showed only minimal amounts of remaining debris and in many specimens only a thin smear layer was detected with many open dentinal tubules.
With the reservation that the comparability of these results obtained in different studies is limited, it can be concluded that obviously, even different rotary nickel- titanium instruments vary in their debris removal efficiency, possibly due to their flute design (Gambarini 1999,Hulsmannet al.2000). For instance, ProFile instruments have radial lands and it has been shown that this file design was less efficient in debris removal compared to rotary instruments having a positive rake angle (Schafer& Zapke2000, Versumer et al.2002). Instruments possessing U-shaped blades with radial land perform a planing action on the root-canal walls rather than a cutting action, which could to be the main reason for the inferior cleaning ability of these instruments (Versumer et al.2002). The K3 files also posses radial lands and this design feature might explain their poorer cleaning efficiency compared to the results obtained with FlexMaster instruments under identical experimental conditions (Schafer & Lohmann 2002).
Summarizing these aspects, it is open to question whether the differences in the cleaning effectiveness of K3 instruments and K-Flexofile observed in the present study has any clinical significance in term of successful canal debridement, particularly as the ability of K3 instruments to maintain the original canal curvature was significantly superior compared with that of K-Flexofiles.
The teeth in all experimental groups were balanced with respect to the apical diameter of the root canal. Furthermore, based on the initial radiograph the teeth were also balanced with respect to the angle and the radius of canal curvature. To achieve this a computerized digital image-processing system was used to determine both the angle and the radius of curvature (Schafer et al. 2002). The homogeneity of the two groups with respect to the defined constraints was examined using a t-test. According to the P-values obtained (Table 1), the groups were well balanced. The curvatures of all root canals ranged between 258 and 358 and the radii ranged between 4.2 and 8.1 mm (Table 1). Thus, the curvatures of the human root canals were comparable to those of the simulated canals in resin blocks used in the first part of this two-part report (curvatures: 288 and 358; radii: 6.5 and 7.5 mm), allowing a comparison of the results obtained in simulated and in human root canals (Schafer & Florek2 002).
Concerning the ability of the two instruments tested to maintain original root curvatures even in severely curved canals, better compliance with original canal shape was obtained using the K3 system compared to hand instrumentation using K-Flexofiles (Fig. 4). A previous study conducted on curved canals in extracted human teeth also demonstrated that the original canal shape was maintained better when using rotary nickel-titanium instruments compared to a hand-preparation technique with stainless steel K-Flexofiles (Bertrand et al. 2001). In general, the results of the present study using extracted human teeth confirm the findings obtained in the first part of this two-part report after preparation of simulated canals (Schafer & Florek in press), in that the use of K3instruments resulted insignificantly less canal transportation than K-Flexofiles. In simulated canals K3 instruments were significantly faster than K-Flexofiles. Certainly, K3 files needed less time to prepare the root canals of real teeth than K-Flexofiles, but this differences was not significant, in contrast to the results obtained in simulated canals.
During the present study, no fractures occurred with K-Flexofiles, whereas five K3 instruments separated, these allwere0.04 taper instruments. Related to the total number of K3 instruments used a fracture rate of approximately 2.1% resulted and related to the total number of real canals enlarged with these instruments a separation rate of approximately 16.7% resulted. Summarizing these data and the findings obtained in the first part of this two-part report after preparation of simulated canals, it seems that during the crown-down technique great caution should be exercised with the use of 0.04 taper K3 files, as already reported for other rotary nickel-titanium instruments (Blum et al. 1999). Certainly, it is important to note that the K3 rotary instrumentation sequence used in our studies is based on the findings obtained in a pilot study, but is not the one recommended by the manufacturer, because up to now, a particular preparation sequence for severely curved canals recommended by the manufacturer were not available. The manufacturer should recommend a well-defined instrumentation sequence for canal preparation which can be modified according to the difficulty of the canal to be enlarged and to the diameter of the original canal in order to minimize the separation rate. The authors acknowledge that the preparation sequence used in the present study may contribute to the separation rate, and therefore, further research seems to be necessary to investigate the influence of different instrumentation sequences on the separation rate of K3 instruments.

References.

American Association of Endodontists (1998) Glossary. Contemporary   Terminology for Endodontics, 6th edn. American Association of Endodontists,   Chicago, Ill, USA.
Bechelli C, Zecchi Orlandini S, Colafranceschi M (1999) Scanning electron   microscope study on the efficacy of root canal wall debridement of hand versus   Lightspeed instrumentation. International Endodontic Journal 32, 484-93.
Bertrand MF, Lupi-Pegurier L, Medioni E, Muller M, Bolla M (2001) Curved   molar root canal preparations using Hero 642 rotary nickel-titanium instruments.   International Endodontic Journal 34,631-6.
Blum JY, Machtou P, Micallef JP (1999) Location of contact areas on rotary   Profile 1 instruments in relationship to the forces developed during mechanical   preparation on extracted teeth. International Endodontic Journal 32, 108-14.
Bolanos OR, Jensen JR (1980) Scanning electron microscope comparisons of   the efficacy of various methods of root canal preparation. Journal of Endodontics   6, 815-22.
European Society of Endodontology (1994) Consensus report of the European   Society of Endodontology on quality guidelines for endodontic treatment.   International Endodontic Journal 27, 115-24.
Gambarini G (1999) Shaping and cleaning the root canal system: a scanning   electron microscopic evaluation of a new instrumentation and irrigation technique.   Journal of Endodontics 25, 800-3.
Gambarini G, Laszkiewicz J (2002) A scanning electron microscopic study of   debris and smear layer remaining following use of GT rotary instruments.   International Endodontic Journal 35, 422-7.
Grandini S, Balleri P, Ferrari M (2002) Evaluation of Glyde File Prepin combination   with sodium hypochlorite as a root canal irrigant. Journal of Endodontics   28, 300-3.
Guerisoli DMZ, Marchesan MA, Walmsley AD, Pecora JD (2002) Evaluation of   smear layer removal by EDTAC and sodium hypochlorite with ultrasonic agitation.   International Endodontic Journal 35, 418-21.
Haikel Y, Allemann C (1988) Effectiveness of four methods for preparing root   canals: a scanning electron microscopic evaluation. Journal of Endodontics14,   340-5.
Hulsmann M (2000) Entwicklung Einer Methodik Zur Standardisierten Uberprufung   Verschiedener Aufbereitungsparameter und Vergleichende in-Vitro-Untersuchung   Unterschiedlicher Systeme Zur Maschinellen Wurzelkanalaufbereitung (Habilitationsschrift).   Gottingen, Germany: University of Gottingen.
Hulsmann M, Rummelin C, Schafers F (1997) Root canal cleanliness after preparation   with different endodontic hand pieces and hand instruments: a comparative SEM   investigation. Journal of Endodontics 23, 301-6.
Hulsmann M, Schade M, Schafers F (2001) A comparative study of root canal   preparation with HERO 642 and Quantec SC rotary Ni-Ti instruments. International   Endodontic Journal 34, 538-46.
Hulsmann M, Versumer J, Schade M (2000) A comparative study of Lightspeed,   ProFile 0.04, Quantec and Hero 642 (Abstract). International Endodontic   Journal 33, 150.
Kochis KA, Walton RE, Lilly JP, Rick's L, Rivera EM (1998) A histology comparison   of hand and Nitto rotary instrumentation techniques. Journal of Endodontics   24, 286 (abstract).
Mizrahi SJ, Tucker JW, Seltzer S (1975) A scanning electron microscopic study   of the efficacy of various endodontic instruments. Journal of Endodontics   1, 324-33. Petschelt A, Stumpf B, Raab W (1987) Tightness of root canal sealers   with and without smear layer. Deutsche Zahnarztliche Zeitschrift 42, 743-6.
Schafer E, Diez C, Hoppe W, Tepel J (2002) Roentgenographic investigation   of frequency and degree of canal curvatures in human permanent teeth. Journal   of Endodontics 28, 211-6.
Schafer E, Florek H (2002) Efficiency of rotary nickel-titanium K3 instruments   compared with stainless steel hand K-Flexofile. Part1. Shaping ability in simulated   curved canals. International Endodontic Journal (in press).
Schafer E, Lohmann D (2002) Efficiency of rotary nickel-titanium FlexMaster   instruments compared with stainless steel hand K-Flexofile. Part 2. Cleaning   effectiveness and instrumentation results in severely curved root canals of   extracted teeth. International Endodontic Journal 35, 514-21.
Schafer E, Zapke K (2000) A comparative scanning electron microscopic investigation   of the efficacy of manual and automated instrumentation of root canals.   Journal of Endodontics 26, 660-4.
Schwarze T, Geurtsen W (1996) Comparative qualitative SEM study of automated   vs. hand instrumentation of root canals. Deutsche Zahnarztliche Zeitschrift   51, 227-30.
SpOngberg L, Engstrom B, Langeland K (1973) Biological effects of dental   materials. Part 3. Toxicity and antimicrobial effects on endodontic antiseptics   in vitro. Oral Surgery 36, 856-71.
Thompson SA, Dummer PMH (1997) Shaping ability of ProFile .04 taper Series   29 rotary nickel-titanium instruments in simulated canals. Part 1 and 2.   International Endodontic Journal 30, 1-15.
Thompson SA, Dummer PMH(1998) Shaping ability of Quantec Series 2000 rotary   nickel-titanium instruments in simulated canals. Part 1 and 2. International   Endodontic Journal 31, 259-74.
Turkun M, Cengiz T (1997) The effects of sodium hypochlorite and calcium   hydroxide on tissue dissolution and root canal cleanliness. International   Endodontic Journal 30, 335-42.
Versumer J, Hulsmann M, Schafers F (2002) A comparative study of root canal   preparation using ProFile .04 and Lightspeed rotary Ni-Ti instruments. International   Endodontic Journal 35, 37-46. West JD, Roane JB, Goerig AC (1994) Cleaning   and shaping the root canal system. In: Cohen S, Burns RC, eds. Pathways   of the Pulps, 6th edn. pp. 179-218. St. Louis, USA: Mosby Year Book.
Wu MK, de Schwartz FBC, van der Sluis WM, Wesselink PR (2001) The quality   of root fillings remaining in mandibular incisors after root-end cavity preparation.   International Endodontic Journal 34, 613-9.
Wu MK, Wesselink PR (1995) Efficacy of three techniques in cleaning the apical   portion of curved root canals. Oral Surgery 79, 492-6.
Yamada RS, Armas A, Goldman M, Lin PS (1983) A scanning electron microscopic   comparison of high volume final flush with several irrigating solutions. Part   3. Journal of Endodontics 9, 137-42.